摘要
针对以人力资源投入为主的传统反窃电模式存在稽查效率低和盲目性等问题,提出一种基于台区识别和关联监测加权算法的窃电检测方法。首先通过用电信息采集系统得到用户基础档案数据,创建台区停电事件校核流程提高台区档案归属准确率。然后采集正常档案用户的疑似窃电异常关联因素,利用改进的层次分析法建立窃电关联监测矩阵模型,实现客户侧窃电智能监控。最后根据置信程度计算用户窃电嫌疑值。通过实例验证提出的算法可以快速准确锁定疑似窃电用户,为反窃电工作提供有力的技术支撑,推动反窃电业务向智能化转变。
Aiming at the problems of low efficiency and blindness in the traditional anti-stealing mode based on human resource investment,a method of electricity stealing detection based on station identification and correlation monitoring weighting algorithm was proposed.Firstly,the user s basic file data was obtained through the electricity information collection system,and the check-up process of the power failure event in the station area was created to improve the accuracy of the file belonging to the station area.Then,the suspected thief-related anomaly related factors of the normal file users were collected,and the improved analytic hierarchy process was used to establish the thief-related correlation monitoring matrix model to realize the intelligent monitoring of the customer side thief.Finally,the user s suspicion value was calculated according to the degree of confidence.An example verified that the proposed algorithm can quickly and accurately lock the suspected thief users,provide powerful technical support for anti-stealing work,and promote the anti-stealing service intelligently.
作者
熊霞
陶晓峰
叶方彬
吴竹筠
XIONG Xia;TAO Xiaofeng;YE Fangbin;WU Zhujun(Nari Group Company Limited/State Grid Electric Power Research Institute,Nanjing Jiangsu 211106,China;Institute of Electric Power Science,State Grid Zhejiang Electric Power Company Limited,Hangzhou Zhejiang 310007,China)
出处
《计算机应用》
CSCD
北大核心
2019年第S02期289-292,共4页
journal of Computer Applications
基金
国家电网公司科技项目(521101180017)
关键词
窃电检测
台区识别
层次分析
关联监测矩阵模型
置信程度
electricity stealing detection
transformer attribute identification
analytic hierarchy process
associated monitoring matrix model
confidence level
